rkernel-densityfilterfunction

Filter function for High Dimensional Data in TDAmapper in R

I would like to use R TDAmapper package to represent my dataset with 76 rows and 316 columns. I'm following this code: http://bertrand.michel.perso.math.cnrs.fr/Enseign but the filter function used, Kernel function kde, is not good for my case because I have a dataset with more then 6 dimensions.

Is there an other Kernel function using like a filter function in TDAmapper for high dimensional dataset? Or Anyone could sugguest me an other filter function?

Thanks in advance

Solution

  • Apply PCA to data, Project your data on Principle components. Say along first PC or etc as a filter